What are the hardware requirements for Distilgpt2 Nepali?

Distilgpt2 Nepali requires approximately 0.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Distilgpt2 Nepali and how large is it?

Distilgpt2 Nepali is developed by Sakonii, a model with 88.2m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Distilgpt2 Nepali?

Distilgpt2 Nepali is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Distilgpt2 Nepali by Sakonii — VRAM 0.3GB

Name: Distilgpt2 Nepali
Author: Sakonii

Distilgpt2 Nepali is an open-source language model by Sakonii. Features: 88.2m LLM, VRAM: 0.3GB, License: apache-2.0, LLM Explorer Score: 0.

Arxiv:1911.02116 Dataset:sakonii/nepalitext-lan... Endpoints compatible Generated from trainer Gpt2 Pytorch Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/Sakonii/distilgpt2-nepali

Distilgpt2 Nepali Parameters and Internals

Model Type

Causal Language Model

Use Cases

Areas:

Nepali text generation

Applications:

Fine-tuning on Nepali language focused downstream tasks

Primary Use Cases:

Nepali text generation

Limitations:

May not perform satisfactorily on shorter sequences due to training on text blocks of size 512 tokens

Supported Languages

Nepali (native)

Training Details

Data Sources:

nepalitext-language-model-dataset, OSCAR, cc100, Wikipedia

Data Volume:

13 million text sequences

Methodology:

Causal language modeling (CLM) objective using Sentence Piece Model (SPM) for text tokenization similar to XLM-ROBERTa

Context Length:

512

Training Time:

around 188.8K training steps

Input Output

Input Format:

text sequence up to 512 tokens

Accepted Modalities:

text

Output Format:

text generation

Performance Tips:

Set a seed for reproducibility due to reliance on randomness in text generation.

LLM Name	Distilgpt2 Nepali
Repository 🤗	https://huggingface.co/Sakonii/distilgpt2-nepali
Model Size	88.2m
Required VRAM	0.3 GB
Updated	2026-07-30
Maintainer	Sakonii
Model Type	gpt2
Model Files	0.3 GB 0.3 GB 0.0 GB
Model Architecture	GPT2LMHeadModel
License	apache-2.0
Model Max Length	512
Transformers Version	4.17.0
Tokenizer Class	XLMRobertaTokenizer
Padding Token	<pad>
Vocabulary Size	50257
Torch Data Type	float32
Activation Function	gelu_new

Rank the Distilgpt2 Nepali capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Distilgpt2 Nepali

Best Alternatives	Context / RAM	Downloads	Likes
Leo DistilGPT Medical	0K / 0.4 GB	15	1
Distilgpt2	0K / 0.4 GB	2759237	634
Distilgpt2 HC3	0K / 0.3 GB	222	1
Distilgpt2 Emailgen V2	0K / 0.3 GB	373	5
Distilgpt2 Emailgen	0K / 0.4 GB	212	4
Distilgpt2 Fables Demo	0K / 0.3 GB	12	0
Distilgpt2 Tiny Conversational	0K / 0.3 GB	386	5

Note: green Score (e.g. "73.2") means that the model is better than Sakonii/distilgpt2-nepali.